Adaptive noise spectral estimation for spectral subtraction speech enhancement
Authors:
- 6 September 2007
- journal article
- Published by Institution of Engineering and Technology (IET) in IET Signal Processing
- Vol. 1 (3) , 156-163
Abstract
An adaptive scheme for noise spectral estimation is proposed to cope with the power spectral subtraction method for speech enhancement. The use of such a noise spectral estimate helps eliminate residual noise without increasing speech distortion. Improvements in the segmental signal-to-noise ratio and modified bark spectral distortion measures are evidently seen with 192 TIMIT sentences corrupted by four types of broadband noise taken from the Noisex-92 database. Moreover, incorporation of this noise adaptation scheme into the constrained parameter estimator is demonstrated. Mean opinion score listening tests confirm that the proposed subtractive scheme yields better results than that acquired by subtracting an averaged noise power spectrum.Keywords
This publication has 23 references indexed in Scilit:
- Speech Enhancement Based onWavelet Thresholding the Multitaper SpectrumIEEE Transactions on Speech and Audio Processing, 2004
- Soft thresholding for DCT speech enhancementElectronics Letters, 2002
- A subspace approach for enhancing speech corrupted by colored noiseIEEE Signal Processing Letters, 2002
- Supplementary schemes to spectral subtraction for speech enhancementSpeech Communication, 2002
- Wavelet speech enhancement based on the Teager energy operatorIEEE Signal Processing Letters, 2001
- Time-domain approach using multiple Kalman filters and EM algorithm to speech enhancement with nonstationary noiseIEEE Transactions on Speech and Audio Processing, 2000
- Iterative and sequential Kalman filter-based speech enhancement algorithmsIEEE Transactions on Speech and Audio Processing, 1998
- Noisy speech enhancement using discrete cosine transformSpeech Communication, 1998
- Suppression of acoustic noise in speech using spectral subtractionIEEE Transactions on Acoustics, Speech, and Signal Processing, 1979
- All-pole modeling of degraded speechIEEE Transactions on Acoustics, Speech, and Signal Processing, 1978